The reliability and validity of the 3-minute all-out cycling critical power test

Research suggests that critical power (CP) can be estimated from a single 3-minute bout of all-out cycling. The purpose of this study was to investigate the reliability and validity of the 3-minute all-out cycling test when carried out at a constant cadence (isokinetic) and against a fixed resistance (linear). Twelve participants completed 8 tests: 1) a ramp test, 2–4) three fixed power tests to calculate CP and Wʹ using the 1/time mathematical model, 5–8) four 3-minute all-out tests to calculate EP and WEP; two isokinetic and two linear. There was no significant difference between EP-isokinetic and CP (P=0.377). There were significant differences between EP-linear and CP (P=0.004), WEP-isokinetic and W′ (P<0.001) and WEP-linear and W′ (P<0.001). Coefficient of variation in EP-isokinetic, EP-linear, WEP-isokinetic and WEP-linear was 1.93%, 1.17%, 8.44% and 5.39%, respectively. The 3-minute all-out isokinetic test provides a reliable estimate of EP and a valid estimate of CP. The 3-minute all-out linear test provides a reliable estimate of EP, but not a valid estimate of CP. Furthermore, these results suggest that the 3-minute all-out test should not be used to estimate Wʹ.